I just had a thought here. Bear with me. With Alpha/NT Digital bore the costs of creating and maintaining the Alpha versions of the OS, tools and applications.
What if AMD did the same thing. Imagine a version of Win2K that was compiled to native K7 code. AMD could re-use the FX!32 stuff that Compaq had put into the kernel. Then they could create a K7-lite that didn't have the x86 baggage. That would be the el-cheapo version. Then you've got the normal K7 that can still run x86 apps.
It would be an interesting experiment and probably scare the crap out of Intel.
Of course, this could also be a crack-pipe fantasy.
P.S. NT on Alpha died when Compaq decided they were pulling the plug. Since Microsoft doesn't sell any hardware, why should they support an architecture that the manufacturer doesn't support anymore?
From the rumors I heard, Intel engineers would be selling their grandmothers to only be using 30 watts. I'll look for the hard numbers, but I've heard _70_ when they run the things at full speed.
Bart: Aw, cheer up, Dad. You make a great hippie. Homer: Aw, you're just saying that. Bart: No, really. You're lazy and self-righteous... Lisa:... and the soles of your feet are jet black!
Before you dismiss this as a troll or a flame, think about it. Twice.
Why discuss American diets and then mention your organic prefs? Is that really on topic. Your point is very well made with only the first and last paragraphs.
You mean they could create a new company. I propose they call it API, for Alpha Processor Inc. That company could create non-Digital/non-Compaq implementations of the 21264. That company could contract with _Intel_ to have their 21264s made, or they could make their own. Or they can get IBM to make them on copper. Or they can get Samsung to make them in Korea.
Even better: They can license their bus to AMD so that motherboards are a commodity and readily available.
I believe Dave Cutler and his team are currently using AXP boxes to do Win64 development while Intel struggles to get their vaporware Itanium/IA-64/Merced crap out the door.
(For those of you hassling MS about Win2K's lateness, don't forget that Merced is 18 months late!)
People keep bringing up the amount of time it took to get C++ standardized as an argument against standardization and standards bodies.
This totally trivializes the situation that happened with C++. Everyone on the committee thought it would be a slam dunk. Then STL comes along, and it was too much of a killer feature not to include it. Furthermore, when the effort started, how many people really had a lot of experience in what was and was not a good idea to put into the standard?
Only that last part would apply to Java.
(P.S. These aren't my original thoughts. I borrowed them from P.J. Plauger's editorial in the latest C/C++ User's Journal)
I didn't see too much about this, so this might or might not be long (depends on how I feel).
I just bailed from an internet startup, and man, have I got some lessons learned (Right, Pat G.?) [Hint: *cough* CBLT *cough*]
I think for quite some time they were serving everything off a 6 way PPro 200, w/ another machine running MySQL. Don't do it!
Version control is your friend. Have a dev, stage and live environment. dev and live doesn't cut it.
If you think you will ever need serious RDBMS capabilities, don't use MySQL. Buy Oracle. Yes it will cost you, but it will be worth not having to migrate from one to the other.
Program for maintainability. You are going to have turnover. You don't want one person to cut crappy code and then leave. You will then have to rewrite that code. If it's worth doing, it's worth doing well.
That's it. I got nothing on network capacity, security, etc. I'm just telling you what I saw, and what I saw that sucked.
I think that the judge understands this. and I'm hoping that he chooses the correct remedy:
(1) Microsoft should be forced to openly publish all of its APIs.
(2) They should be prohibited from utilizing any API that has not been completely disclosed.
(3) They should be forced to provide correction, clarification and explanation, when the behavior of any API differs from the published specification, or is unclear or appears to be in error.
(4) They should be forced to disclose all new APIs as they are created, to facilitate third parties' efforts to make their software compatable with Microsoft software.
[end quote] How many people that come up with these ideas program for windows? Let's take, as an example, the MFC. Microsoft publishes an API for those. They also provide you with the _source code_. Now we all know that access to the source is a good thing. You can figure out what the author(s) was(were) thinking. But, MS will not guarantee the implementation will remain the same. Why should they? Think. Why are there private member variables in good object oriented languages?
I think if you go back and look at this, you'll see that the OS folks don't create undocumented APIs for the Office folks. It's the other way around. The joes (and janes) who write Office come up with a cool feature that goes into mso??.dll. If it's cool and useful, then the OS folks grab it for use in the os.
Disclaimer: If someone shows me the prototype for BreakNetscapeReallyBadEx in Win98SE, I'll stand corrected.
You are somewhat correct when you state that IE is not free for use in a business environment. Hey, you gotta pay someone to install it:)
Seriously, MS will _give_ you the IEAK for under $10. All they ask in return is a quarterly statement of how many installs you are doing. Not free, but not the $30-40 that NetScrape is used to charging in that market.
In a word, no. Are there some things that suck? Sure. ActiveX documents have been dropped like a dirty sock. But the actual WebBrowser component works like a charm. It is extremely simple to create your own version of IE (GreenRDExplorer?) with VB or VC++. I can't admit to having done it with VJ++ tho;)
Why do you say it's not good food? Wheat is wheat, right?
I'd also disagree with you that the gov't is helping farmers. More like helping ADM.
At the end of the day, should hundreds of folks live at (or below) subsistence level because their farming implements consist of a wooden stick and some muscle power? Or should that one dude in Kansas just ride his combine listening to Alabama and grow their food for them so they can get down to the serious business of making wiring harnesses for cars, or airplanes, or whatever?
Some places have an advantage at producing certain goods. Don't fight that. If the US can produce good food cheap, let them. Then all those local farmers in 3rd world countries can get down to doing something they can make money at.
I don't understand how you can say their dev tools are overpriced. Have you compared the price of Visual C++ to Sun's visual environment?
I also that that one could argue that MS Office is so expensive because of the capabilities it offers. To my knowledge none of the other Office competitors allow the same extensibility and a simple, unified development language.
Of course, that's not to say that you can't get the job done with vi/emacs and Perl and all the shell tools.
Every rule has an exception, especially in very large corporations likely to have these debate-inspiring "no OSS" policies.
Approved by whom? To what extent (written,verbal,implicit understanding)? How does the approval get recorded for use during an ISO 9000 audit? Does this jive with our system architecture plans?
These are all things that large companies do (very broad generalization:)
You'll save a lot of ink if you say "No software that IT (or MIS) doesn't install. In God we trust, all others bring facts and data."
I look at the thread, and the subject appears to be why would a company have a "no OSS" policy?
However, the thread quickly degenerated into "I use Perl and it works, NT sucks, blah blah blah"
There were a few posts that mentioned what is the probable reason (IMHO) for the no OSS policy. A no OSS policy means that you have to get approval to install software. This lets legal check the license and avoid any problems where someone installs something that is not freeware in a corporate setting (Applixware anyone?).
BTW, the subject was a poke at those who quickly took this thread off topic. Why not post some relevant comments about how to create a business case to get OSS software approved?
You left out another category where cottage industries can't compete. Need a hint? Mosey over to your local airport. See that big metal tube that holds 300+ people? Ah, now you've caught my point.
There are some systems that are so complex and require so much specialized knowledge that a cottage industry can't compete. I for one would never fly on a commercial aircraft produced by cottage industry.
This has nothing to do with producing mass quantities either. How many Concordes did the British and French build?
This does have some applicability to software. Joe Hacker cutting code in his basement isn't going to have the resources to perform usability testing or reach a non-computer literate audience. That's going to require the resources of a corporation. Try licensing Snoopy or some other piece of intellectual property. Again, a corporation (unless you happen to be a lawyer/hacker).
Slashdot Mirror
I just had a thought here. Bear with me.
With Alpha/NT Digital bore the costs of creating and maintaining the Alpha versions of the OS, tools and applications.
What if AMD did the same thing. Imagine a version of Win2K that was compiled to native K7 code. AMD could re-use the FX!32 stuff that Compaq had put into the kernel. Then they could create a K7-lite that didn't have the x86 baggage.
That would be the el-cheapo version. Then you've got the normal K7 that can still run x86 apps.
It would be an interesting experiment and probably scare the crap out of Intel.
Of course, this could also be a crack-pipe fantasy.
P.S. NT on Alpha died when Compaq decided they were pulling the plug. Since Microsoft doesn't sell any hardware, why should they support an architecture that the manufacturer doesn't support anymore?
From the rumors I heard, Intel engineers would be selling their grandmothers to only be using 30 watts. I'll look for the hard numbers, but I've heard _70_ when they run the things at full speed.
From a Simpson's episode:
... ... and the soles of your feet are jet black!
Bart: Aw, cheer up, Dad. You make a great hippie.
Homer: Aw, you're just saying that.
Bart: No, really. You're lazy and self-righteous
Lisa:
Before you dismiss this as a troll or a flame, think about it. Twice.
Why discuss American diets and then mention your organic prefs? Is that really on topic. Your point is very well made with only the first and last paragraphs.
What about fluorinated drinking water?
Clearly a communist plot. Have you ever seen a communist drink fluorinated water? No, they drink grain alcohol and rainwater.
Peace on Earth through Purity of Essence.
You mean they could create a new company. I propose they call it API, for Alpha Processor Inc.
That company could create non-Digital/non-Compaq implementations of the 21264. That company could contract with _Intel_ to have their 21264s made, or they could make their own. Or they can get IBM to make them on copper. Or they can get Samsung to make them in Korea.
Even better: They can license their bus to AMD so that motherboards are a commodity and readily available.
Oh, wait. All that's happened!
FYI, the technology that InHell is using to run 32 bit x86 stuff on Merced was licensed from Digital/Compaq (think FX!32).
Who's going to be better at emulation/translation? The folks who've been working on it for years, or the licensee?
I'm sure someone will correct me if I'm wrong:
I believe Dave Cutler and his team are currently using AXP boxes to do Win64 development while Intel struggles to get their vaporware Itanium/IA-64/Merced crap out the door.
(For those of you hassling MS about Win2K's lateness, don't forget that Merced is 18 months late!)
God, I hope so. I'd hate for the FCC to bust in and confiscate my pair of them...
I'm looking at the e-speak license, and to me, at least, it looks like the license Sun should have used for Java.
:)
GPL, LGPL, they keep the trademark on the name, and you get to use the trademark if you're compatible.
If I'm way out of line, tell me. I really want to know. (Unless you're gonna tell me to pour hot grits down my pants
People keep bringing up the amount of time it took to get C++ standardized as an argument against standardization and standards bodies.
This totally trivializes the situation that happened with C++. Everyone on the committee thought it would be a slam dunk. Then STL comes along, and it was too much of a killer feature not to include it. Furthermore, when the effort started, how many people really had a lot of experience in what was and was not a good idea to put into the standard?
Only that last part would apply to Java.
(P.S. These aren't my original thoughts. I borrowed them from P.J. Plauger's editorial in the latest C/C++ User's Journal)
That almost sounds like another product I've
heard of.
I can't remember the company, but they just implemented a script runtime component, and you can plug your script interpreter into it.
(For the sarcasm impaired, the product is IE. And yes, PerlScript does exist for it.)
This is hilarious! I can just hear that
Phil Hartman/'Troy McClure' voice.
Now, if only the original poster had used their
name...
I didn't see too much about this, so this might or might not be long (depends on how I feel).
I just bailed from an internet startup, and man, have I got some lessons learned (Right, Pat G.?)
[Hint: *cough* CBLT *cough*]
I think for quite some time they were serving everything off a 6 way PPro 200, w/ another machine running MySQL. Don't do it!
Version control is your friend. Have a dev, stage and live environment. dev and live doesn't cut it.
If you think you will ever need serious RDBMS capabilities, don't use MySQL. Buy Oracle. Yes it will cost you, but it will be worth not having to migrate from one to the other.
Program for maintainability. You are going to have turnover. You don't want one person to cut crappy code and then leave. You will then have to rewrite that code. If it's worth doing, it's worth doing well.
That's it. I got nothing on network capacity, security, etc. I'm just telling you what I saw, and what I saw that sucked.
[quote]
I think that the judge understands this. and I'm hoping that he chooses the correct remedy:
(1) Microsoft should be forced to openly publish all of its APIs.
(2) They should be prohibited from utilizing any API that has not been completely disclosed.
(3) They should be forced to provide correction, clarification and explanation, when the behavior of any API differs from the published specification, or is unclear or appears to be in error.
(4) They should be forced to disclose all new APIs as they are created, to facilitate third parties' efforts to make their software compatable with Microsoft software.
[end quote]
How many people that come up with these ideas program for windows? Let's take, as an example, the MFC. Microsoft publishes an API for those. They also provide you with the _source code_. Now we all know that access to the source is a good thing. You can figure out what the author(s) was(were) thinking. But, MS will not guarantee the implementation will remain the same. Why should they? Think. Why are there private member variables in good object oriented languages?
I think if you go back and look at this, you'll see that the OS folks don't create undocumented APIs for the Office folks. It's the other way around. The joes (and janes) who write Office come up with a cool feature that goes into mso??.dll. If it's cool and useful, then the OS folks grab it for use in the os.
Disclaimer: If someone shows me the prototype for
BreakNetscapeReallyBadEx in Win98SE, I'll stand corrected.
Dude! Get your facts straight before you
;)
speak. Otherwise, you might seem like you're
not as educated as you make yourself out to be.
Corsair == GM, not Ford.
While some 70s Mustangs (read Mustang II) might have had a flamability problem, I think you should look at the base platform, the Ford Pinto.
Why no mention of the flammable 80's Chevy trucks?
P.S. Whose law is it that says mentioning 'Nazi' loses you the argument?
Sounds like COM in Windows. I can specify
,anyone?)
what program is going to handle a specific
URL (via it's interface, TweakUI, or the registry).
I can even add my own protocols
(biteme://
You are somewhat correct when you state that IE is not free for use in a business environment. Hey, you gotta pay someone to install it:)
Seriously, MS will _give_ you the IEAK for under $10. All they ask in return is a quarterly statement of how many installs you are doing. Not free, but not the $30-40 that NetScrape is used to charging in that market.
So you've heard, eh?
;)
In a word, no. Are there some things that suck?
Sure. ActiveX documents have been dropped like a dirty sock. But the actual WebBrowser component works like a charm. It is extremely simple to
create your own version of IE (GreenRDExplorer?)
with VB or VC++. I can't admit to having done it with VJ++ tho
Think for yourself or don't think at all.
Why do you say it's not good food? Wheat is wheat, right?
I'd also disagree with you that the gov't is helping farmers. More like helping ADM.
At the end of the day, should hundreds of folks live at (or below) subsistence level because their farming implements consist of a wooden stick and some muscle power? Or should that one dude in Kansas just ride his combine listening to Alabama and grow their food for them so they can get down to the serious business of making wiring harnesses for cars, or airplanes, or whatever?
What's going to grow the world's economy more?
Don't you remember the lesson from Econ 101?
Some places have an advantage at producing certain goods. Don't fight that. If the US
can produce good food cheap, let them. Then all those local farmers in 3rd world countries can get down to doing something they can make money at.
I don't understand how you can say their
dev tools are overpriced. Have you compared
the price of Visual C++ to Sun's visual environment?
I also that that one could argue that MS Office is so expensive because of the capabilities it offers. To my knowledge none of the other Office competitors allow the same extensibility and a simple, unified development language.
Of course, that's not to say that you can't get the job done with vi/emacs and Perl and all the shell tools.
Every rule has an exception, especially in
:)
very large corporations likely to have these
debate-inspiring "no OSS" policies.
Approved by whom? To what extent (written,verbal,implicit understanding)? How does the approval get recorded for use during an ISO 9000 audit? Does this jive with our system architecture plans?
These are all things that large companies do (very broad generalization
You'll save a lot of ink if you say "No software that IT (or MIS) doesn't install. In God we trust, all others bring facts and data."
I look at the thread, and the subject appears
to be why would a company have a "no OSS" policy?
However, the thread quickly degenerated into
"I use Perl and it works, NT sucks, blah blah blah"
There were a few posts that mentioned what is the probable reason (IMHO) for the no OSS policy. A no OSS policy means that you have to get approval to install software. This lets legal check the license and avoid any problems where someone installs something that is not freeware in a corporate setting (Applixware anyone?).
BTW, the subject was a poke at those who quickly
took this thread off topic. Why not post some relevant comments about how to create a business case to get OSS software approved?
You are in a twisty maze of passages, all alike.
Plover!
You left out another category where cottage industries can't compete. Need a hint? Mosey over to your local airport. See that big metal tube that holds 300+ people? Ah, now you've caught my point.
There are some systems that are so complex and require so much specialized knowledge that a cottage industry can't compete. I for one would never fly on a commercial aircraft produced by cottage industry.
This has nothing to do with producing mass quantities either. How many Concordes did the British and French build?
This does have some applicability to software. Joe Hacker cutting code in his basement isn't going to have the resources to perform usability testing or reach a non-computer literate audience. That's going to require the resources of a corporation. Try licensing Snoopy or some other piece of intellectual property. Again, a corporation (unless you happen to be a lawyer/hacker).